Fifteen rats, fed Diet 2000 ad libitum, were inoculated with Actinomyces viscosus T-6. Once the rats harbored a manifest flora at Actinomyces they were fed by gastric intubation and were not allowed to eat Diet 2000. The level of infection was monitored once every week by oral swabbings which were primarily stained with FA. During a period of two weeks the prevalence of Actinomyces decreased rather slowly and it was decided that four animals should be given headcups to prevent them from eating feces which could eventually affect the oral flora. Also, at the same time, six other animals were given casein ad libitum, which in a previous study had proved not to support implantation of Actinomyces in the rat. The rest of the rats were continually fed just by gastric intubation. In all rats in the experiment the Actinomyces flora decreased below detectable levels, which occurred within different periods of time depending on the dietary regimen applied. However, characteristically high numbers of Actinomyces were rapidly recovered once the rats were again fed a carbohydrate containing diet ad libitum.